Accelerating scientific discovery

July 2023

Overview

2

• About eLife (very quickly)
• Limitations of the current publishing system
• eLife new publishing model

​

​

elifesciences.org

accelerating scientific discovery

About eLife

3

• Established in 2012, an independent nonprofit led by scientists
• Publishes reviewed preprints in the life and medical sciences
• Develops open-source platforms to support the review, organisation and dissemination of research
• Promotes openness, integrity and equity, diversity and inclusion
• Tasked from our funders to change scientific publishing for the better

elifesciences.org

accelerating scientific discovery

4

• Increase science visibility and dissemination
• Stimulate the advancement of science
• Provide quality control on what is published (e.g., by peer review)
• Reduce the gap between knowledge and practice/policy making

​

Median Franco and Lopez and Lopez 2022; Brainard 2021; Roche et al., 2021

What is the purpose of Scientific Publishing?

elifesciences.org

accelerating scientific discovery

The current system is slow

The average time to publication is 9 months – about the same time as human gestation

5

Bjork and Solomon, 2013

Christie et al., 2021

elifesciences.org

accelerating scientific discovery

6

Current metrics (e.g., citations and impact factor) do not capture individuals’ scientific impact.

The validity of citation-based metrics (including the impact factor) is being compromised.

Davies et al., 2021

Fire and Guestrin 2019

The current publishing metrics are not reliable

elifesciences.org

accelerating scientific discovery

The current system under-values the time of academics

7

• Time spent on peer reviews that get lost when the paper �is rejected
• Time spent reformatting papers

Puehringer et al., 2021; Aczel et al., 2021

elifesciences.org

accelerating scientific discovery

8

Costs keep increasing in the current system

Evidence suggests supra-inflation increase of both subscription prices for scholarly journals and open access publishing fees.

Mekonnen et al., 2021

Van Noorden 2013

Grossmann and Brembs, 2021

elifesciences.org

accelerating scientific discovery

9

Scientists want change…

Survey by eLife in 2021 showed a demand for an evolution in publishing by early career researchers, especially transparency

​

​

elifesciences.org

accelerating scientific discovery

10

…and there is broader movement for change

• White House Office of Science and Technology Policy:
• Establish strategic approaches for advancing �open science
• Promote equitable participation in open science through transparency, integrity and equity of reviews
• Account for open science activities in evaluations and incentives
• Engage underrepresented communities in the advancement of open science and research

elifesciences.org

accelerating scientific discovery

Preprints are a first step in a new system

• “A preprint is a version of a scientific manuscript posted on a public server prior to formal peer review” 
• “As soon as it’s posted, your preprint becomes a permanent part of the scientific record, citable with its own unique DOI”

Source: PLOS survey https://plos.org/open-science/preprints/ 

11

elifesciences.org

accelerating scientific discovery

Preprints are widely accepted in grant applications

12

elifesciences.org

accelerating scientific discovery

Preprints – Pros and cons

13

Advantages

• Immediate access to research
• Preprints are free to deposit and access
• Controlled by authors

Disadvantages

• Have not been reviewed
• Not curated

​

elifesciences.org

accelerating scientific discovery

In a world where preprints are the norm, what is the role of journals?

14

elifesciences.org

accelerating scientific discovery

Our criteria for a better model

15

Increased value

Faster

Good for �science

Good for �scientists

More control �to authors

More �transparent

elifesciences.org

accelerating scientific discovery

16

The new publishing model at eLife

• Provides expert public review and assessment of preprints
• Promotes scientists’ evaluation based on what, not where, they publish
• The version of record is akin to a traditional journal article

elifesciences.org

accelerating scientific discovery

17

Submission (1)

• We review only preprints
• Editors (experts in their field) consult and commit to review preprints where we have the required expertise and can add most value with public review

elifesciences.org

accelerating scientific discovery

18

Peer Review (2)

• We no longer issue accept/reject decisions after peer review
• Editors and reviewers consult to write public reviews and the eLife assessment that summarizes the significance of the work and the strength of the evidence
• Authors can correct factual errors prior to publication

elifesciences.org

accelerating scientific discovery

19

Publication (3)

• Every preprint reviewed by eLife is available on the eLife’s website as a Reviewed Preprint
• This includes the paper, the eLife assessment, the public reviews and (if provided) the initial authors’ response
• All Reviewed Preprints have a DOI and are citable

elifesciences.org

accelerating scientific discovery

20

How it looks

eLife assessment: curation that sits under the Abstract (difficult to miss)

​

Indication that the work is a reviewed preprint

Public peer reviews available within the eLife website

Article review history

elifesciences.org

accelerating scientific discovery

21

eLife assessments: common vocabulary

elifesciences.org

accelerating scientific discovery

eLife assessment: examples

22

elifesciences.org

accelerating scientific discovery

23

Author Revision (4)

• Authors can submit a revised manuscript for re-review at any time following the first peer review
• After re-review, a new Reviewed Preprint is produced with updated reviews and eLife assessment

elifesciences.org

accelerating scientific discovery

24

Version of Record (5)

• At any time after a Reviewed Preprint is published, authors can declare it the Version of Record (VOR)
• VOR declaration marks the end of the review process and triggers production of the version sent for indexing in databases such as PubMed and Web of Science

elifesciences.org

accelerating scientific discovery

25

The new eLife model: overview

elifesciences.org

accelerating scientific discovery

Our criteria for a better model

26

Increased value

Faster

Good for �science

Good for �scientists

More control �to authors

More �transparent

elifesciences.org

accelerating scientific discovery

Benefits: faster

• Preprints posted immediately
• 1-3 month publication time for Reviewed Preprints

​

​

​

27

Better value

Faster

Good for �science

Good for �scientists

More control�to authors

More �transparent

elifesciences.org

accelerating scientific discovery

Benefits: good for science

• Richer, yet still compact, assessments for use in hiring and funding decisions
• Quicker publication can lead to faster scientific progress
• Transparency increases scrutiny

​

​

28

Faster

Good for �science

Good for �scientists

More control�to authors

More �transparent

Better value

elifesciences.org

accelerating scientific discovery

Benefits: good for scientists

• Published output aligned with funder requirements
• Visible quality indicators (e.g., eLife assessment)
• Promotes scientists’ evaluation based on what, not where they publish

29

Faster

Good for �science

Good for �scientists

More control�to authors

More �transparent

Better value

elifesciences.org

accelerating scientific discovery

Benefits: more transparent

• Assessment criteria visible
• Peer comments published
• Author comments published
• Versions tracked

​

​

30

Faster

Good for �science

Good for �scientists

More control�to authors

More �transparent

Better value

elifesciences.org

accelerating scientific discovery

Benefits: more control to authors

• Authors decide on VOR
• Authors choose how to respond to reviews
• Authors choose to make amendments

​

​

31

Faster

Good for �science

Good for �scientists

More control�to authors

More �transparent

Better value

elifesciences.org

accelerating scientific discovery

Benefits: better value

• Publishing fee reduced to $2,000
• Full waivers are available
• Values the time of authors and reviewers

​

​

​

32

Faster

Good for �science

Good for �scientists

More control�to authors

More �transparent

Better value

elifesciences.org

accelerating scientific discovery

Our action contributes to debate and discussion

33

elifesciences.org

accelerating scientific discovery

More information

34

https://elifesci.org/eLife-New-Model

elifesciences.org

accelerating scientific discovery

35

Public support

eLife's funders as well as other funding bodies and research institutions support the use of reviewed preprints in research assessment

University of Bristol Library Services

elifesciences.org

accelerating scientific discovery

Common questions (for reference)

36

Follow @eLife��Visit elifesciences.org

elifesciences.org

accelerating scientific discovery

Will my research still count?

• Yes, moving to VOR is akin to a traditional Journal article
• Journal title is often used as signal for quality and evaluation. We propose a new system where each paper is evaluated individually, based on the eLife assessments we provide (with full expert public reviews also included)
• Everything else is the same:
• High quality reviews
• Published article with a pdf version
• High editorial standards
• Press releases and promotion

37

elifesciences.org

accelerating scientific discovery

How are you choosing what to review?

• We are capacity-limited and cannot review everything. We will review papers where we have relevant expertise and where we think public reviews and assessments will add most value
• Scientists active in the relevant field decide whether or not to review a preprint
• Vast majority of preprints is seen by multiple experts before deciding whether to review or not
• We will monitor our decisions and take corrective action where needed

38

elifesciences.org

accelerating scientific discovery

What if authors get bad reviews?

• We make sure reviews are constructive
• Authors have 2 weeks (flexible) to raise any concerns before we post the reviews and the eLife assessment, and they can also write an initial response to the reviews (not meant to be a revised paper at this stage)
• Reviews will be posted for all articles we send to review (as a Reviewed Preprint)
• We reserve the right not to publish anything that can cause public harm, and/or a threat beyond what can directly concern human health and safety (we predict this to be very rare)

39

elifesciences.org

accelerating scientific discovery

Can authors take the reviewed preprint and publish it somewhere else?

• Yes, they can
• The VOR is akin to a journal article, therefore other publishers cannot consider this for publication in their Journals

40

elifesciences.org

accelerating scientific discovery

Will peer reviews be signed?

• Peer reviews will be signed by eLife, as organisation, rather than by the reviewers themselves. We will make sure reviews are fair and constructive
• Reviewers can post their named reviews on bioRxiv or other database where the paper is

​

41

elifesciences.org

accelerating scientific discovery

Will eLife pay reviewers?

• Not at the moment. We are aware of the value added by reviewers and in discussion to develop ways to recognise their work
• eLife is open to the possibility of paying reviewers. As a non-profit organisation, our aim is to support the community and sustain our publishing model, not increase our revenue. Thus, paying reviewers would likely result in increasing the publishing fee

42

elifesciences.org

accelerating scientific discovery

Will eLife Impact Factor change?

• eLife does not officially recognise the use of the impact factor in research evaluation and does not track it
• The new eLife publishing model promotes research evaluation based on what is published in each individual article and not on the impact factor of the journal where the article is published

43

elifesciences.org

accelerating scientific discovery

Time investment from the editors

• We do not expect editors to spend significantly more time than before handling papers and deciding if to review them
• If the workload increases for editors at any stage of the process, we will take corrective action and recruit more editors

44

elifesciences.org

accelerating scientific discovery

Review Commons: what is it?

• A platform that provides peer review of preprints before submission to a journal
• Selected manuscripts are first reviewed to output a reviewed preprint. Reviews and the authors’ response appear alongside the reviewed preprint on the preprint database
• The reviewed preprint can then be submitted to one of the Review Commons affiliate Journals for publication (traditional publishing system)
• A valuable service, however different from what eLife is doing (next slide)

45

elifesciences.org

accelerating scientific discovery

eLife and Review Commons: main differences

• The eLife publishing process includes initial preprint assessment and peer review, eventual re-review and re-assessment, and VOR production. Review Commons provides review of the preprint; all steps after that are done by the affiliate Journal and incur a post-review accept-reject decision
• Review Commons provides no equivalent to the eLife assessment in their reviewed preprints

​

46

elifesciences.org

accelerating scientific discovery

Peer Community In (PCI): what is it?

• Organization of researchers offering peer review, recommendation and open access publication of preprints
• Following submission by the authors, a “recommender” (researcher in a field) selects preprints that are reviewed by experts
• Following peer review, the recommender issues an accept/reject decision on the reviewed preprint
• If accepting, the recommender writes a “recommendation” based on the peer reviews summarising the significance and solidity of the findings
• Authors can publish their reviewed preprint in another Journal

47

elifesciences.org

accelerating scientific discovery

eLife and PCI: main differences

• At PCI, the “recommender” (expert in a given field) issues accept/reject decisions after preprint review
• At PCI, only accepted preprints are citable and receive the so called “recommendation” written by the recommender
• Reviews are published only for accepted preprints
• The PCI recommendation (equivalent of the eLife assessment) is typically more complex and longer
• The PCI recommendation is written by one expert only

48

elifesciences.org

accelerating scientific discovery

F1000 Research: what is it?

• A publishing platform offering open peer review of all submitted manuscripts
• Following editorial checks for adherence to policies, all submitted articles are published
• Open peer review is then conducted on all published articles and peer reviews are published alongside the authors’ responses
• All reviewed articles are labelled with one of 3 statuses (i.e., “Approved” “Approved with reservations” and “Not approved”)
• Authors can revise their paper and have its status updated
• Papers successfully passing peer review (depending on the status) are indexed

49

elifesciences.org

accelerating scientific discovery

eLife and F1000: main differences

• At eLife an extensive board of scientists (>800) decide what preprints are peer reviewed and oversee/organise the peer review and assessment process
• eLife produces a citable output, the Reviewed Preprint, that can be used in Research Assessment *(as publicly stated by different funders)
• The eLife assessment written by experts to reflect with a common vocabulary the strength of the data and the significance of the findings brings article-level curation. It is distinct from the accept/reject-like decisions “Approved”, “Approved with reservations” and “Not approved” of reviewed papers at F1000

​

​

50

elifesciences.org

accelerating scientific discovery

What about front matter articles?

• For the time being, Magazine articles, Review Articles and Scientific Correspondence do not need to be posted as preprints, and remain subject to accept/reject decisions after review

51

elifesciences.org

accelerating scientific discovery

Retractions

• If a Reviewed Preprint is found to be fraudulent or deeply flawed, we will retract it and ask the authors to withdraw the preprint from the preprint server
• If the article in question has already moved to VOR, it will also be retracted (along with the preprint and all Reviewed Preprint versions) in line with the existing workflow for retractions

52

elifesciences.org

accelerating scientific discovery

How do I cite a reviewed preprint?

For bibliographies:

Jordan Rebecca, Keller Georg B. (2023) The locus coeruleus broadcasts prediction errors across the cortex to promote sensorimotor plasticity eLife 12:RP85111

https://doi.org/10.7554/eLife.85111.1 - uses “version DOI” that refers to the specific article version (e.g., reviewed preprint, VOR, etc.) being cited

​

For CVs:

Jordan Rebecca, Keller Georg B. (2023) The locus coeruleus broadcasts prediction errors across the cortex to promote sensorimotor plasticity eLife 12:RP85111

https://doi.org/10.7554/eLife.85111 - uses “parent DOI” that always refers to latest version of the article

This important study provides convincing evidence that locus coeruleus is activated during visuomotor mismatches. Gain of function optogenetic experiments complement this evidence and indicate that locus coeruleus could be involved in the learning process that enables visuomotor predictions. This study therefore sets the groundwork for the circuit dissection of predictive signals in the visual cortex. Loss-of-function experiments would strengthen the evidence of the involvement of locus coeruleus in prediction learning. These results will be of interest to systems neuroscientists.

53

elifesciences.org

accelerating scientific discovery